1 / 20

Presenter Hari Surineni Dec 6, 2005 COMP -7760 , Prof Stan Franklin

Shakey - The Robot Stanford Research Institute AIC Program:   Funding:  Rome Air Development Center Contract No.  AF 30(602)-4147 Performance Period:  1966-03-17 to 1968-12-05. Presenter Hari Surineni Dec 6, 2005 COMP -7760 , Prof Stan Franklin. Shakey.

hoang
Download Presentation

Presenter Hari Surineni Dec 6, 2005 COMP -7760 , Prof Stan Franklin

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Shakey - The RobotStanford Research Institute AIC Program:  Funding:  Rome Air Development CenterContract No.  AF 30(602)-4147Performance Period:  1966-03-17 to 1968-12-05 Presenter Hari Surineni Dec 6, 2005 COMP -7760 , Prof Stan Franklin

  2. Shakey Named for its erratic and jerky style of movement, Shakey stands six feet tall and is equipped with a TV camera, a triangulating range finder, bumpers, and a wireless video system.

  3. Shakey • Developed by Stanford Research Institute Artificial Intelligence Center from 1966 through 1972 , Shakey has had a substantial legacy and influence on present-day artificial intelligence and robotics.

  4. Shakey - Features • Shakey was the first mobile robot to be controlled by programs that reasoned. • The major aim of the Shakey project was to investigate the use of logic-based problem solving. • The control of movement and the interpretation of sensory data was secondary.

  5. Sensors, Environment, Motivation and Effect Environment • Office environment with specially colored and shaped objects Motivation • To navigate from one room to another and move boxes

  6. Sensors, Environment, Motivation and Effect Sensors • Videocon TV camera • Optical range finder • Whisker bump sensors • Radio link to DEC PDP-10 , PDP-15 computers • Some the above parts have been modified in SHAKEY II, III, IV &V

  7. Parts • Antenna • Range finder • TV camera • On board logic • Camera control unit • Bump detector • Caster wheel • Drive motor • Drive Wheel

  8. Shakey - Tasks • A typical task for Shakey was to find a box of a given size, shape and colour in various rooms and to then move it to a designated position. • To accomplish this, it had to be able to cope with obstacles and plan its actions • http://www.ai.sri.com/shakey/images.php

  9. Shakey • To recognize objects in its world, the MIT blocks world program was adapted. • Environment had to be tailored to the limitations of this system. • Only objects that could be recognized were ones which were simple, uniformly colored and flat-faced. Even the rooms had to be specially constructed.

  10. Shakey • To reason, Shakey used STRIPS (Stanford Research Institute Problem Solver). • Given a task, a plan is constructed out of a limited set of allowable actions of the form: Preconditions --> Action --> Consequents. • This is very similar to theorem proving. • Classical AI algorithms such as A* search were used.

  11. Shakey • Commands are given as “action routines” and very abstract . For example , • Command to Go_Thru (D1,R2,R1) means • “Go from room R1 to room R2 via doorway D1”

  12. Shakey • Shakey had a limited ability to handle unexpected outcomes. • Each move took 1 hour of computing time and had a high probability of failure. • STRIPS produced good plans but the sensory interpretation of the environment was just too primitive

  13. Shakey - Architecture

  14. Shakey - Architecture • Shakey used programs for perception, world-modeling, & acting. • Low level action(LLA) routines took care of simple moving, turning, and route planning. (Represented by black lines in the figure – fast LLAs) • Intermediate level actions (ILA) strung the low level ones together in ways that robustly accomplished more complex tasks. ILAs represent simple, symbolic knowledge based actions. (broken grey arrows)

  15. Shakey - Architecture • High level actions (HLAs – grey arrows) represent complex symbolic knowledge based reasoning, such as plans. The highest level programs could generate and execute plans to achieve goals. • The system also generalized and saved these plans for possible future use.

  16. Shakey - Conclusion • Considering the time period it was built, Shakey’s vision recognition software to distinguish between boxes, walls and doorways was very advanced. • The vision software converted the TV image into a binary image and found a number of features, such as the area and perimeter of connected regions, and the position and orientation of those regions in the visual field.

  17. Shakey - Conclusion • Despite limitations , like slowness of its execution cycle, Shakey was a success. It performed the required tasks. • It created a deeper understanding of problems associated with mobile, embodied agents. • Gave significant insights on future development of robotic architecture. One of the most important lessons learnt was creation of computational world models was expensive.

  18. Shakey • SHAKEY now resides in the Computer History Museum, and has been inducted into the Robot Hall of Fame. • To view the historical Shakey documentary, visit http://www.ai.sri.com/movies/Shakey.ram http://www.ai.sri.com/autogen/video_list.php

  19. Questions? Happy holidays

More Related